Attempts have been made to reduce the incidence of postdural puncture headache (PDPH) after spinal anesthesia by changing the size and design of the needle. We wished to determine whether these strategies are effective in reducing PDPH and whether they affect the incidence of back pain and the failure rate of spinal anesthesia. The literature was searched for trials comparing noncutting spinal needles with cutting needles and larger spinal needles with smaller needles. Trials were included if they were randomized or blinded and if outcomes included PDPH, backache, or failure of the method. The pooled odds ratio for each side effect was computed, and the results were considered statistically significant if the 95% confidence interval excluded 1. Four hundred fifty articles were identified by title using computerized search strategies. Thirty-one abstracts, 25 correspondences, 44 original articles, and 12 reviews were assessed. There was a reduction in the incidence of PDPH when noncutting spinal needles rather than cutting needles were used (P < 0.05), unless the discrepancy in needle size was very large. There also was a reduction in PDPH when a small spinal needle was used compared with a large needle of the same type (P < 0.05). There was no difference in the incidence of failure of spinal anesthesia or the incidence of back pain. We conclude that a noncutting needle should be used for patients at high risk for PDPH, and the smallest gauge needle available should be used for all patients.